The Puzzle of Alzheimer's: Finding the Pieces
Neuroscientists and clinical investigators at all three Mayo campuses provide a powerful synergy, empowering one of the top Alzheimer’s disease research collaborations in the country. From imaging engineers to biochemists to physician specialists, Mayo Clinic is pursuing the answers from every angle. Three young researchers have broken new ground in their part of that quest. By taking a novel approach and screening for unique coding in the genome, they’ve come across particular variants that may influence development of dementia differently in individuals.
Trying to understand Alzheimer’s disease is like trying to put together a huge jigsaw puzzle, with no idea of what the picture should look like in the end. Just as no one piece can solve the puzzle, no single gene causes Alzheimer’s disease. In fact, researchers have already looked at more than 600 genes in over 1000 studies, many with contradictory findings. Now Mayo Clinic is taking a new approach to uncover genes that previously eluded scientists. It is also trying pinpoint which gene variants are likely to put people at risk — or even protect them — from developing the disease.
“We believe that even though many people with Alzheimer’s disease might look like they have the same disease, the factors behind their illness are actually quite different,” said Fanggeng Zou, Ph.D., a postdoctoral researcher in the lab of Steven Younkin, M.D., Ph.D., at Mayo Clinic in Florida. “That is why the research is taking so long, because you have to find every factor — small factors and big factors alike — and understand how they can combine to produce the disease.”
A Welcome Challenge
Not to be deterred by this complexity, Dr. Zou and colleagues Minerva Carrasquillo, Ph.D., and Nilufer Taner M.D., Ph.D., were motivated to look for genetic variants in a completely different way. Variants exist naturally in our individual genomes and are one of the things that make each human unique. Genetically speaking, humans are just 99.9 percent identical, which means that our nucleotide sequence — the order of our A’s, G’s, T’s and C’s, — varies at about one out of every 1000 nucleotides. Most variants are innocuous, but a small proportion can change the structure of the protein product or alter expression levels — that is, the “turning on” or “off” of the gene. The Mayo researchers began their investigation by hunting for variants associated with such changes in gene expression levels. Their approach was a stark departure from previous investigations, which had searched for variations associated with the single outcome — of either of having or not having Alzheimer’s.
“With the disease approach, you have a yes or no type of outcome that can be very limiting,” says Dr. Taner, who has her own laboratory. “Our studies instead take as the outcome something that you can quantify, and as a result the type of data generated is often much richer once you go out of the box of only thinking about the disease.” Their untraditional tactics paid off, resulting in two back-to-back papers in the journals Neurology and PloS ONE in the last year. In the Neurology study, Zou tested the researchers’ approach by measuring expression levels of a dozen genes that had previously been implicated in Alzheimer’s. The group found that variants in or around a particular gene (which Dr. Taner first linked to Alzheimer’s as a post-doctoral fellow) influenced its level of expression in the brain.
In the Plos ONE study, Dr. Carasquillo screened the entire length of this gene — called insulin-degrading enzyme or IDE — for sequence changes that impacted gene expression. The variant that emerged was, quite surprisingly, one inextricably connected to the top contender identified from the previous study. Thus, two different approaches led researchers to the same gene and the same variants.
“The thing is, no one would have guessed a priori that these variants would have done anything,” said Carasquillo, a postdoctoral fellow in Dr. Younkin’s lab.
Fitting it Together
But the notion that this gene in particular could play a role in Alzheimer’s was not a surprise. A number of studies have definitively shown that IDE can break down amyloid beta, the protein that clumps together in the brains of Alzheimer’s patients. And though no one knows the underlying cause of this devastating form of dementia, the prevailing theory does center on a buildup of amyloid beta protein, resulting in brain cell death and memory loss.
The gene variants discovered in the Mayo research tie in nicely to that theory. The variants are protective, meaning they turn on the expression of IDE, producing more of the protein to clear out amyloid beta. Dr. Carasquillo says when they looked at levels of amyloid beta in brain samples and drew associations with the presence of the variant, it all made sense. The levels of this “bad” protein went down when the gene variant was present.
On its own, this variant cannot predict whether someone will get Alzheimer’s. In fact, it is unlikely that clinicians will ever be able to definitively tell someone that they will develop the disease, says Dr. Carasquillo. But the variant does add to a growing list of protective and risk factors — genes, amyloid beta levels, MRI results, cognitive scores — that one day could be pieced together to give patients the whole picture of their disease risk.
No one knows exactly what causes Alzheimer’s disease. Experts know that beta-amyloid proteins malform into clumps called plaques, which appear to damage neurons in the brain. Neurons also appear to be subject to tangles of an essential protein called tau. Many researchers believe these tau tangles also damage the neurons, causing them to die. The exact mechanisms and relationships of these apparent causes are unclear. It is clear that the five million Americans affected by this dementia will more than triple in coming decades as Boomers age.
“Complex diseases are very hard to diagnose just based on genetic risk factors,” says Dr. Carasquillo. “Right now, the main goal of research in this field is to find what genes are involved, not just to assess risk but also because all of them are potentially therapeutic targets. If you find a gene that is involved in the disease you may be able to figure out how to treat a patient using either the product of that gene or altering the levels of that gene somehow. Hopefully, by knowing as many genes as possible that are involved in the disease, we can devise new ways to prevent it, treat it, or diagnose it better.”
The Mayo researchers have expanded their initial studies as far as possible, measuring expression levels of all the genes in two different brain regions of people with and without Alzheimer’s. They are investigating how variants in those genes correlate with or influence their expression levels. With around 24,000 genes in the human genome, it is a huge project, but one that could yield critical information for clinician-investigators like Dr. Taner.
“Alzheimer’s disease is very challenging because it is not treatable,” she says. “But day after day, I hear the family members of my patients say that I have told them things that have helped them deal with the illness. They are thankful for the information they receive, so it gives me an understanding that even just being informed about the disease can be helpful. So as long as I am able to convey new information, even though I am not able to give them a cure as yet, that gives me a sense of satisfaction.”
— Marla Broadfoot, Ph.D., July 2010